Customer Support Alert: Flirting with a Chatbot? Cease the Interaction Immediately - Understanding the Boundaries of AI Assistance

Customer support chatbots assist users in resolving daily issues across various areas, such as online shopping and financial services. However, they can also become targets for flirtatious customers. A recent survey indicates that nearly 20% of Americans have flirted with chatbots.

Image courtesy of Pixabay

Cybersecurity experts warn that a significant desire for human contact and a focus on quickly resolving issues could lead to identity theft and severe privacy concerns. "Many users are still unaware that when they seek customer support, they are often interacting with a chatbot, not a human. People flirt with chatbots out of curiosity, confusion, loneliness, and even for sexual reasons.

This behavior raises psychological and privacy concerns. Individuals may compromise their privacy by divulging personal data to resolve issues or to receive emotional support," explains Adrianus Warmenhoven, a cybersecurity expert at NordVPN.

Pushing yourself off a privacy cliff

Engaging in flirtatious conversations with chatbots can be risky in terms of digital privacy. Customers often disclose more personal information to the chatbot than necessary, aiming to impress an imaginary person on the other end.

Similarly, customers may inadvertently reveal sensitive personal data, such as their ID or social security number, when they are eager to quickly resolve an issue. However, the chatbot's responses may repeatedly ask them to "rephrase the question" or "provide more details about the problem." This situation highlights the potential dangers of sharing too much information with chatbots.

While privacy issues are not necessarily the concern here, it's important to note that all data entered into a chatbot is collected, stored, and accumulated. Any system, including chatbots, can have vulnerabilities—flaws, gaps, or unintentional "backdoors"—that hackers can exploit, especially if the chatbot does not adequately protect customer data using encryption.

"Customer support operators used to act as a filter, understanding the domain and privacy risks and requesting only relevant and less sensitive information. Now, AI must understand nuances in what people say they need versus what they actually need. While turning to AI for support functions is inevitable, consumers will bear more responsibility for deciding what data to share with a chatbot. They must be extra cautious about the information they disclose, as they cannot predict how this information will be used in the future, particularly since it is sometimes used for training algorithms," says Warmenhoven.

How to protect your privacy from customer support chatbots

To protect your privacy while using chatbots, Adrianus Warmenhoven suggests these preventive measures:

1. Only provide the information necessary to resolve the issue. Avoid flirting with chatbots or sharing highly personal information that you wouldn't want to become public in case of a leak.
2. When drafting a request message, avoid including any information that could identify you or others. Use specific identifiers like an order number for online shops or a booking number for flight tickets. This information should be sufficient for identification. Avoid providing sensitive information such as ID numbers, social security numbers, or bank card numbers. Also, refrain from signing your message with your full name, as this is not necessary.
3. Prepare your request and information before approaching the customer support chatbot. Drafting a message in advance in a notepad app allows you to review the message for clarity and ensure you provide only the necessary information.
4. To protect your identity from cybercriminals, always request a verification email from the chatbot. This is a common practice and an effective tool. Reputable businesses often offer this function in their privacy protection measures.